A5035811403- Medical Image Segmentation Techniques
- Radiomics and Machine Learning in Medical Imaging
- Medical Imaging Techniques and Applications
- Advanced X-ray and CT Imaging
- Lung Cancer Diagnosis and Treatment
- Advanced Neural Network Applications
- Advanced MRI Techniques and Applications
- Computer Graphics and Visualization Techniques
- Image Retrieval and Classification Techniques
- Ultrasound Imaging and Elastography
- Ultrasound and Hyperthermia Applications
- 3D Shape Modeling and Analysis
- AI in cancer detection
- Advanced Radiotherapy Techniques
- Context-Aware Activity Recognition Systems
- EEG and Brain-Computer Interfaces
- Medical Imaging and Analysis
- Photoacoustic and Ultrasonic Imaging
- Cardiac Imaging and Diagnostics
- Renal cell carcinoma treatment
- Advanced Vision and Imaging
- Augmented Reality Applications
- Video Surveillance and Tracking Methods
- Robotics and Sensor-Based Localization
- Cerebrovascular and Carotid Artery Diseases
Université de Rennes
2015-2024
Laboratoire Traitement du Signal et de l'Image
2015-2024
Inserm
2012-2024
Centre National de la Recherche Scientifique
2022
Hôpital Civil, Strasbourg
2022
Southeast University
1993-2021
Université Rennes 2
2014
Institut de Recherche en Santé, Environnement et Travail
2009
Knowledge of left atrial (LA) anatomy is important for fibrillation ablation guidance, fibrosis quantification and biophysical modelling. Segmentation the LA from Magnetic Resonance Imaging (MRI) Computed Tomography (CT) images a complex problem. This manuscript presents benchmark to evaluate algorithms that address segmentation. The datasets, ground truth evaluation code have been made publicly available through http://www.cardiacatlas.org website. also reports results Left Atrial Challenge...
Renal cancer is one of ten most common cancers in human beings. The laparoscopic partial nephrectomy (LPN) becomes the main therapeutic approach treating renal cancer. Accurate kidney and tumor segmentation CT images a prerequisite step surgery planning. However, automatic accurate remains challenge. In this paper, we propose new method to perform precise angiography images. This relies on three-dimensional (3D) fully convolutional network (FCN) which combines pyramid pooling module (PPM)....
This paper describes a fast and fully automatic method for liver vessel segmentation on computerized tomography scan preoperative images. The basis of this is the introduction 3-D geometrical moment-based detector cylindrical shapes within minimum-cut/maximum-flow energy minimization framework. represents an original way to introduce data term as constraint into widely used Boykov's graph cuts algorithm, hence, automate segmentation. evaluated compared with others synthetic dataset. Finally,...
Abstract Background Renal cancer is one of the 10 most common cancers in human beings. The laparoscopic partial nephrectomy (LPN) an effective way to treat renal cancer. Localization and delineation tumor from pre-operative CT Angiography (CTA) important step for LPN surgery planning. Recently, with development technique deep learning, neural networks can be trained provide accurate pixel-wise segmentation CTA images. However, constructing training dataset a large amount annotations...
Accurate segmentation of uterus, uterine fibroids, and spine from MR images is crucial for high intensity focused ultrasound (HIFU) therapy but remains still difficult to achieve because 1) the large shape size variations among individuals, 2) low contrast between adjacent organs tissues, 3) unknown number fibroids. To tackle this problem, in paper, we propose a kernel Encoder-Decoder Network based on 2D model. The use can capture multi-scale contexts by enlarging valid receptive field. In...
Image registration is fundamental in medical imaging, enabling precise alignment of anatomical structures for diagnosis, treatment planning, image-guided interventions, and longitudinal monitoring. This work introduces IMPACT (Image Metric with Pretrained model-Agnostic Comparison Transmodality registration), a novel similarity metric designed robust multimodal image registration. Rather than relying on raw intensities, handcrafted descriptors, or task-specific training, defines semantic...
This paper presents a three-dimensional edge operator aimed at the detection of anatomical structures in medical imaging. It uses spatial moments gray level surface, and operates three dimensions with any window size. allows us to estimate location contrast as well surface orientation. The computation discrete version is reported. Bias errors due sampling noise are analyzed both theoretical experimental level. moment-based compared other well-known operators on simple shaped primitives for...
Many questions remain open in most of the physics-based therapies that we have been shortly reviewed. All are based on principles discovered a long time ago, and advances still far from being sufficient. Clinical studies do not clarify enough how when they should be used first intention or sequenced over time. In other words, patients must cared today tomorrow. There is no doubt these offer many opportunities. New probe transducer design device miniaturization through microelectromechanical...
The security of elderly people living alone is a major issue. A system that detects anomalies can be useful for both individual and retirement homes. In this paper, we present an adaptive human tracking method built on particle filter, using depth thermal information based the velocity t he position head. main contribution paper fusion to improve tracking. For each frame, there new combination coefficients weighting. Results show deal with cases fast motion (fall), partial occultation scale...
This paper describes a new method for the three-dimensional (3-D) tracking and quantification of blood vessels from magnetic resonance angiography (MRA). The approach is based on 3-D geometrical moments consists following steps: (1) interactive selection seed points; (2) autom atic vessels; (3) local computation both diameter orientation; (4) rendering vessels. detection estimation scheme has been validated simulated real data.
Abstract Learning embryology remains difficult, since it requires understanding of many complex phenomena. The temporal evolution developmental events has classically been illustrated using cartoons, which create difficulty in linking spatial and aspects, such correlation being the keystone descriptive embryology. We synthesized bibliographic data from recent studies atrial septal development. On basis this synthesis, consensus on stages septation as seen human heart reached by a group...
Image denoising and signal enhancement are the most challenging issues in low dose computed tomography (CT) imaging. Sparse representational methods have shown initial promise for these applications. In this work we present a wavelet based sparse representation technique utilizing dictionary learning clustering. By using wavelets extract suitable features images to obtain accurate atoms algorithm. To achieve improved results also lower number of clusters which reduces computational...
In this paper, we address the particularly challenging problem of calibrating a stereo pair low resolution (80 × 60) thermal cameras. We propose new calibration method for such setup, based on sub-pixel image analysis an adequate pattern and bootstrap methods. The experiments show that achieves robust with quarter-pixel re-projection error optimal set 35 input pairs pattern, which namely outperforms standard OpenCV procedure.